IBM, Pfizer develop AI to predict Alzheimer's disease

The duo explained that the AI model’s natural language processing allowed them to pick up subtleties and changes in speech that they may have otherwise been missed.

IBM and Pfizer have developed an artificial intelligence (AI) model that can help predict Alzheimer’s disease in healthy people before symptoms develop.

The model uses short language sample from a standard verbal cognitive test which asks people to look at a drawing and say what they see.

According to IBM, the AI model gives predictive feedback that has an accuracy of 70%, which is more accurate than predictions made from other available biomedical data at 59%.

“Our vision is that one day clinicians will have multiple AI and machine learning tools to help identify if an individual is at risk of developing Alzheimer’s disease,” IBM said in a release.

The duo explained that the AI model’s natural language processing allowed them to pick up subtleties and changes in speech that they may have otherwise been missed.

The data samples used for making the predictions were from a group of healthy people who eventually did or did not develop the disease later in life, allowing researchers to verify the accuracy of the AI model’s prediction, IBM noted.

As Alzheimer’s disease can affect a wide spectrum of individuals, the researchers also focused on assessing the risk of the disease in the general population, and not just focused on high-risk individuals with a genetic history or predisposition to the disease, it added.

OCTOBER 31, 2020

https://www.thehindu.com/